International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
Medremind: A Mobile Phone Based Medicine
In-take Helper
Nikhil Khobragde, Prashant Kalbande, Durgesh Nagalwade, Ms. Prerana jaipurkar
Student, Department of Information Technology, SRPCE, Nagpur, India
ABSTRACT
Out-patient medication administration has been identified as the most error-prone procedure in modern healthcare. Under or
over doses due to erratic in-takes, drug-drug or drug-food interactions caused by un-reconciled prescriptions and the absence of
in-take enforcement and monitoring Mechanisms have caused medication errors to become the common cases of all medical
errors. Most medication administration errors were made when patients bought different prescribed and over-the-counter
medicines from several drugs stores and use them at home without little or no guidance. Medremind can remind its users to
take the correct medicines on time and Record the in-take schedules for later review by healthcare professionals. Medremind
has two distinguished features: (1) it can alert the patients about potential drug-drug/drug-food interactions and plan a proper
in-take schedule to void these interactions; (2) it can revise the in-take schedule automatically when a dose was missed. In both
cases, the software always tries to produce the simplest schedule with least number of intakes. Medremind is equipped with
user friendly interfaces to help its users to recognize the proper medicines and obtain the correct instructions of taking these
drugs. It can maintain the medicine in-take records on board, synchronize them with a database on a host machine or upload
them onto a Personal Heath Record (PHR) system. A proof-of-concept prototype of Medremind has been implemented on
Window Mobile platform and will be migrated onto Android for Google Phones
Keywords- Telemonitoring, Medication Error, Mobile Computing, Time scheduling.
1. INTRODUCTION
According to a landmark study on medical errors conducted by the US Institute of Medicine in 1999, medication Errors
(ME) and adverse drug reactions (ADR) are the most common cases among all medical errors. These adverse Drug
events (ADE) incurred significant tolls in terms of patient fatality, financial costs (including additional medical
Expenses, lost income and productivity), and damages to the reputation and morale of healthcare professionals. Most of
these errors are nonetheless preventable. One study found 530,000 preventable ADEs among Medicare out patients
each year. Although errors can occur in every step of medication process during medicine prescription, dispensing and
administration, they happen most frequently in the prescription and administration stages. In the past decade,
increasingly wide-spread use of computerized physician order entry (CPOE), clinical decision support systems (CDSS)
and electronic medical records (EMR) along with better procedures to dispense medicine has helped to eliminate a
large proportion (up to 80%) of prescription errors, which account for half of all medication errors. In comparison,
little progress has been made in the prevention of administration errors, which are caused by improper and failed use of
prescribed medicine. Consequently, medicine administration errors have become the prevalent cause of ADEs. They
accounted for 25%–40% of all medication errors and were the main reason for admission of elderly into nursing homes.
Out-patient medication administration has been identified as the most error prone procedure amidst the entire
medication process.
Most of these errors were made when patients bought different prescribed and over-the-counter (OTC) medicines from
several drug stores and use them at home without little or no guidance. Common causes of these errors include:
irregular medicine in-takes due to the patient’s busy or erratic lifestyles, complicated in-take schedules due to many
medicines and doses taken by the patient, adverse drug reactions caused by un-reconciled prescriptions obtained from
different sources, lack of knowledge about proper use of medicines, (5) lack of consultation with healthcare providers
when confusion arises and lack of monitoring mechanisms to keep track of patient’s medicine in-take. Recently,
telemedicine, especially telemonitoring techniques, has been investigated as a cost-effective approach to control quality
of care (QoC) in out-patient medication administration. By sending in-take reminders to the patient (even producing
the proper medicine from a medicine dispenser) and then recording patient’s responses, Health Maintenance
Organizations (HMO) hope to reducing cost of service while improving quality of care. Communication between HMOs
and patients is established through wired or wireless Internet connections. Although these efforts represent progress in
the right direction, the medicine dispensers thus made are bulky, expensive and prone to dispensing errors. In this
paper, we introduce Medremind, a smart phone application designed to help patients to avoid medicine administration
Volume 2, Issue 3, March 2013
Page 361
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
errors. The software is named after the “Eye of Egyptian God Horus” [insert] from which the prescription symbol Rx
was derived. Medremind can perform the following three primary functions:
 Issue medicine in-take reminders
Medremind will issue an alert approximately 5 – 15 minutes (preset by user) before the scheduled time to take certain
medicine(s). The alert will be issued repetitively until it is cancelled by the user. Scheduling of in-take alerts is
performed by a real-time process/resource scheduling algorithm that can satisfy time constraints according to medicine
in-take directions and drug-drug/drug-food interactions. This function is integrated with the calendar and planner
applications installed on most smart phones.
 Provide medicine identification and in-take directions
Medremind has a built-in database containing crucial information about the medicines (including their photo images,
in-take directions and precautions) and the healthcare providers (including physicians, pharmacists and HMOs)
relevant to its user. All these data can be retrieved with the touch of a button while Medremind is in use.
 Maintain medicine in-take records
Medremind will record the time at which its user cancels an in-take alert and regard that at the time that specific
medicine(s) was taken. These medicine in-take records can be stored on board, synchronize with the database on a host
machine and/or uploaded onto a Personal Heath Record (PHR) system.
Medremind has two distinguished features: (1) it can alert the patients about potential drug-drug/drug-food interactions
and plan a proper in-take schedule to avoid these interactions; (2) it can revise the in-take schedule automatically when
a dose was missed. In both cases, the software always tries to produce the simplest schedule with least number of intakes. A proof-of-concept prototype of Medremind has been implemented on the Window Mobile platform and will be
migrated onto Android for Google Phones.
2. RELATED WORK
As mentioned in the introduction, most attempts to reduce medicine administration errors have focused their efforts on
developing “medicine dispensers”. Most commercial products are low cost, manual operating devices. A weakness
shared by these devices is that their users must load medicine doses into these devices and then program their
operation. Naturally, such a cumbersome operation is susceptible to human errors. Automated medicine dispensers are
definitely a step taken in the right direction. Some of these devices will be installed in the homes for the elderly or
chronically ill. Medremind represents a novel attempt to integrate healthcare support with mobile computing. As next
generation smart phones such as iPhone™ and Google Phones™ become popular, such an application has the potential
to reap a huge market share.
3. SYSTEM OVERVIEW
THE SYSTEM ARCHITECTURE OVERVIEW INCLUDING MOBILE PHONE WITH MEDREMIND, APACHE SERVER, APACHE
APPLICATION INTERFACE AND ENCRYPTED DATA STORE, HOSPITAL, PHARMACY AND OTHER MEDICAL SERVICES. THE
MOBILE PHONE IS THE END POINT OF THE SYSTEM AND COMMUNICATES WITH USER DIRECTLY.
Medremind on mobile phone uses wireless to connect to Indivo Server and then download medicine scheduling
specification as the input of Medremind. Medremind schedules the medicines in-take according to MSS and writes the
whole in-take schedule to Calendar application on and notes down user’s medicines in-take records and uploads the intake records to Indivo. If doctors or pharmacists change the content of prescription such as increase dosage of some
medicines, the Medremind can download the latest version of MSS from Indivo and reschedule. Indivo is a distributed,
web-based, personally controlled health records (PCHR) system for information exchange and communication. It
follows the open standard PCHR and provides API is accessible on the Internet. The patient can easily take his own
medical treatment records. Pharmacies and hospitals are the data provider of the system. They upload medical records
to Apache and view or down load medical records only when they have user’s permission. So that they can really now
and control the patient’s condition and improve the medical service quality.
Volume 2, Issue 3, March 2013
Page 362
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
Figure 1: Medremind System Architecture
4. OPERATION SCENARIOS
After Medremind writes medicines in-take schedule to Calendar, it set the every medicine in-take as a calendar event.
The Calendar will pop out a message 15 minutes before medicine in-take time to reminder user. If user clicks OK,
Medremind will take down the records. When user delays or miss some doses, Medremind checks the schedule
presently is feasible or not and readjust the schedule and rewrites to Calendar. If feasible schedule is impossible,
Medremind sends Warning message to user to tell him (her) the situation is urgent or not. If urgent, it suggest user to
make contact with doctor immediately.
Figure 2: Mss production process
 Medication Prescription Compilation
The data flow in the production of Medication Scheduling Specification (MSS). A patient may be take care by several
physicians, each of whom issues prescription using a different CPOE system. Tools provided by Personal Health Record
Systems such as Indivo may be used to transcribe medicine prescriptions and in-take directions to standard format and
then merged to produce a medication scheduling specification (MSS) for a specific patient. The MSS XML file will be
sent to Medremind via RSS subscription.
 Medication Scheduling Specification
Table 1 shows the scheduling specification of a medicine. Such a specification consists of three parts: Prescription
Parameters (PP), Dosage Parameters (DP) and Interaction Parameters (IP).
Table 1: Sample Medication Scheduling Specification
Prescription Parameters (PP)
Medicine Identifier
M
Medicine Dose
g
Medicine Form
Capsule/Tablet/..
Medicine Amount
n
Therapy Duration
T
Dosage Parameters (DP)
Nominal Minimum and
Volume 2, Issue 3, March 2013
Page 363
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
Maximum Dose
Nominal Minimum and
Maximum Separations
Maximum Intake B
over interval R
(B, R)
Minimum Intake L over
interval P
(L, P)
Absolute Minimum and
Maximum Dose
[Dmin, Dmax]
Absolute Minimum and
Maximum Separations
[asmin, asmax]
Interaction Parameters (IP) <List>
Interferer Identifier
N
Minimum Separation from M
to N
minToInterferer
Minimum Separation from N
to M
min FrInterferer
 Prescription Parameters (PP)
The first part gives information Medremind needs to know about the medicine including its name or identifier M and
the duration that the patient shall take this medicine. The medicine comes in dose g (medicine dose); dose size
parameters of the medicine are given in terms of integer multiples of g. The medicine form indicates the unit of dosage,
such as capsule, tablet…etc. The total amount of medicine patient need to in-take is medicine amount. The part also
provides other relevant attributes such as a picture image of the medicine for verification purpose. The Horus uses the
same time resolution for all medicines. All separation parameters expressed are in terms of multiples of Horus time
resolution. We use one hour hereafter unless stated otherwise.
 Dosage Parameters (DP)
The dosage parameters part specifies constraints on dose size and separation (i.e., the length of time interval between
any two consecutive doses) for scheduling the medicine when the medicine is taken alone. Take the direction of Advil
for example: a part of it reads “Take 1 gel caplet every 4 to 6 hours. If pain or fever does not respond to 1 capsule, 2
capsules may be used.” So, its nominal dose size and separation ranges are [1,2] and [4,6], respectively. Supply Rate
(B, R) of M: It indicates that the intake (i.e. the total size of all doses) within any time interval of length R must be no
more than B. For example, the supply rate of Advil is (6,24) because its direction also says “Do not exceed 6 gel
capsules in 24 hours. Demand Rate (L, P) of M: It indicates that the intake within any interval of length P must be at
least L. Many medicines (e.g., antibiotic and insulin) have demand rate constraint to ensure that at least the minimum
required amount is at work.
 Interaction Parameters (IP)
We refer to a medicine or food that interacts with M to the extent as to require some changes in how M is to be
administered as an interferer N of M. The IP section of M contains an entry for each of its interferers. The entry of an
interferer N may also define additional separation constraints, each of which specifies a required time separations
between each dose of M and any dose of the interferer N. Table 1 lists only the minimum separation min To Interferer
from the medicine to interferer for each dose M scheduled before any dose of N and the minimum separation min FrI
interferer from the interferer to the medicine for each dose of N scheduled before some dose of M. Take Fosamax as an
example. This medicine for prevention and treatment of brittle bone decease must be taken on empty stomach, and the
user should not take anything within 30 minutes after taking the medicine. Hence the minimum separation parameters
to and from any interferer of Fosamax are half an hour and 6 hours, respectively. In this the required separations
between doses of interferers make scheduling more difficult.
Medicine In-Take Scheduling Algorithms
This section provides an overview of the algorithms for scheduling multiple medicines. The algorithms work with fixed
dose sizes. As stated earlier, that a valid dose size exists has already been assured when the user’s MSS was generated.
By first choosing a valid dose size for each medicine, the scheduler then focuses on finding times before individual
doses to meet all intra-medicine and inter-medicine separation constraints.
 Scheduling Models
1) Resource Model
Volume 2, Issue 3, March 2013
Page 364
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
The design of the scheduler is based on the resource model that uses a virtual processor PM and a virtual resource RM
for each medicine M to keep track of when the user is available to take the medicine. When computing a schedule, the
scheduler treats each dose of each medicine M as if it is a job on processor PM and the sequence of doses of the
medicine as a task M. A job starts when the corresponding dose should be retrieved by the user. A schedule for the
medicine is a list of the time instants at which jobs of task M start.
2) Resource Allocation Rules
The scheduler maintains correct separations between doses of M by scheduling each of the corresponding jobs non
preemptively on PM for this amount of time whenever possible, but it may schedule the job for a smaller amount time
in the range from the absolute minimum separation asmin(M) of M to nsmin(M). In addition, the maximum absolute
separation between consecutive doses of M is enforced by imposing a relative deadline for each of the jobs. For each
medicine M that has interferers, the scheduler uses the virtual resource RM and resources of the interferers to help it
maintain inter-medicine separations. Simply put, a job of M can start at a time t only when the resource RM is free at t.
The scheduler allocates the resource RM to each job of N for minFrInterferer units of time each time when it schedules
a job on PN. Thus, each job of the interferer blocks jobs of M from starting for this amount of time. The worst-case
blocking time of the medicine M is the maximum over all of its interferers of the minimum separations from interferer
to M.
3) Priority Schemes
Both OMAT and ODAT algorithms operate based on priorities. One of the better priority schemes is the Most
Victimized First (MVF) scheme which gives priorities to tasks based on their worst case blocking times; the longer the
worst case blocking time, the higher the priority. Other priority schemes based on separation and interference
characteristics of the medicines include the Most Interferers First (MIF) scheme and the Shortest Separation Difference
First (SSDF) schemes. They give higher priorities to tasks corresponding to medicines with larger numbers of
interferers or larger differences between the maximum and minimum nominal separations, respectively. We also
experimented with the classical real-time priority schemes Rate Monotonic (RM) and Earliest Deadline First (EDF)
schemes. The former gives higher priorities to tasks with shorter periods. The latter gives priorities to jobs based on
their absolute deadlines; the earlier the deadline, the higher the priority. Clearly, EDF scheme is suitable for ODAT
algorithms only.
 Scheduling Algorithms
It provides the pseudo-code description of basic OMAT and ODAT algorithms. The inputs are MSS of all medicines
and Priority Schemes used by the algorithms. The Boolean output variable feasible indicates whether the algorithm
succeeded in finding a feasible schedule when it terminates. If it succeeded (i.e., feasible is TRUE), the elements of the
feasible Schedule array point to the schedules i.e., the lists of job start time of all medicine in the MSS.
1) One-Medicine-at-A-time (OMAT)
After creating instances of data structures described above, the scheduler considers medicines in non-increasing priority
order: It schedules individual dose of each medicine M as a separate job, starting from the first job until it either fails to
find an available (start) time for some job of M before the duration T of the medicine or has successfully generated a list
of start times for all the jobs within the duration. An available time for a job is an instant between the release time and
deadline of the job at which both PM and RM is free. The search for the earliest of such instants and the bookkeeping
chores of the scheduler are described by. In the former case, the scheduler returns immediately with feasible set to
FALSE. In the latter case, it sets the element in feasibleSchedule[] for the medicine to point to the newly generated list
and then move on to work on the medicine next in priority order.
One-Dose-at-A-Time (ODAT)
The scheduler does the same work when it schedules individual jobs regardless of the algorithm it uses. An ODAT
scheduler assigns priorities to jobs according to PrioritySchemes when the jobs are released. It uses the array
releaseTimesCurrentJobs[] to keep track when the current job of each medicine is released and ready to be scheduled.
The schedule of a medicine M is complete when the possible start time of the job of M currently being scheduled is
later than the minimum duration of M. The scheduler continues to schedule jobs as they are released in priority order
until either it fails to find a possible start time for the job currently being scheduled or the schedules of all medicines in
the given MSS are complete.
5. IMPLEMENTATION OF IN-TAKE REMINDER AND MONITOR
Volume 2, Issue 3, March 2013
Page 365
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
This section documents the implementation details of Medremind including its platform, software design and user
interfaces.
 System Platform
Medremind is developed on .net framework with windows mobile 6.0 SDK and integrate with Calendar application
through the Calendar application interface released by Microsoft. After Medremind schedule all medicines, the system
writes the medicine in-take schedule to Calendar application. Medremind inputs are user references and MSS of all
medicines including medicine directions and user prescriptions. Medremind download MSS of all medicines taken by
its user from an Indivo personal health record (PHR) server. In addition, user set his/her preferences such as sleep time
and busy time using the Medremind graphic user interface. If feasible, the scheduler will avoid scheduling medicine intakes when the user is busy or resting. Medremind outputs include full in-take schedule, revised medicine in-take
schedule and medicine in-take records. After Medremind create a feasible full in-take schedule, it writes the schedule to
the Calendar application through an application programming interface (API). The Calendar application plays the role
of reminder and interacts with user directly. If the user miss or delay some dose, the system will readjust the in-take
schedule and rewrite to Calendar application. The readjust schedule is revised medicine intake schedule. When the user
clicks OK to confirm the dosage in-take, the system takes down it as medicine in-take records.
 Software Design
Figure 3: Medremind Input/output flows
Figure 3 shows the implementation architecture. The Medremind downloads the MSS of user in-take medicines from
Indivo. The user sets the preference and then starts the medication scheduler. If the Medication Scheduler creates
feasible schedule successfully, it shows some message to graphic user interface and writes the schedule to Calendar
application. After that the Medication Scheduler monitors the user take medicine on time or not and readjust the intake
schedule.
Figure 4: Medremind Functional Modules
The class diagrams of Medremind are classified to three categories: Medication Schedule Specification, Medicine
Priority and Medication Scheduler. As the Figure 4 shows, the data structure of MSS includes Dosage Parameters,
Interaction Parameters and Interferer Parameters. It holds all related constraints values. The Medicine Priority defines
five schemes of priority including EDF, MIF, MVF, RM and SSDF. The EDF is used by ODAT only. Medication
Scheduler is the core of Medremind. It creates a Job Model object, a Resource Model object and a Schedule object for
each medicine. The Job Model object holds the values of jobs parameters of each medicine M. The execution Time field
is set to the nominal minimum separation nsmin of M initially. The Resource Model maintains two integer arrays
resource and processor. The initial values of every element of the two arrays are 0 which indicates the resource and
processor are free. When a job is scheduled on processor, the processor is set to 1 for k elements in processor array
which indicates the processor is occupied by the job for the length of execution time k. The resource of M is occupied
Volume 2, Issue 3, March 2013
Page 366
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
by the interacting medicines or food N of M. Only when the processor and M are free for the length of execution time
and the resource is free at the start instance of job, the job can be scheduled and executed. The Schedule array is the
outputs of Medication Scheduler to show the full in-take schedule. The Medication Scheduler supports three kinds of
scheduling algorithms including ODAT, OMAT with as soon as possible and OMAT with as fast as possible. The other
methods in Medication Scheduler are private methods used for Medication Scheduler only. The graphic user interface
designed in Medremind version.
Figure 5: Medication Scheduler Classes
 User Interfaces
The graphic user interface we design for Medremind version.1. The first page shows three buttons for user to set
preference or load a MSS file or display the result schedule and a text show the next dose user have to intake. The
second one shows the preference setting page. The user can set the unavailable time or special time which they don’t
want to take medicine.
6. SNAPSHOTS
Snapshot 1: Application Startup
Snapshot 2: Medicine Scheduler
Snapshot 3: Alert Dialog
7. CONCLUSIONS
In this paper, we presented the design ideas behind Medremind, a medicine in-take reminder and monitor installed on
a smart phone. This mobile computing application combines mobile phone based telemonitoring techniques with realtime scheduling algorithms to offer ubiquitous services to numerous out-patients. Notable accomplishments include the
development of OMAT/ODAT medication scheduling algorithms and the implementation of an integrated mobile
computing application. Plenty of work remains to be done after this initial effort: first and foremost, a thorough
integration of Medremind with electronic medical records (eMAR) and electronic personal health records (ePHR).
Advance scheduling algorithms including those that can produce incremental changes to existing schedules should be
investigated. The possibility of exercising on-the-fly changes of prescriptions in response to patient’s conditions may
also be explored.
Volume 2, Issue 3, March 2013
Page 367
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 3, March 2013
ISSN 2319 - 4847
REFERENCES
[1] Koppel B, et al., “Role of Computerized Physical Order Entry Systems in Facilitating Medication Errors,” Journal
of American Medical Association, Vol. 293, No.10, 2005
[2] Kuperman GJ, Bobb A, Payne TH, et al. “Medication Related Clinical Decision Support in Computerized Provider
Order Entry Systems: A Review” Journal of American Medical Informatics Association, 2007.
[3] Baron RJ, Fabens EL, Schiffman M, Wolf E, “Electronic Health Records: Just around the corner or over the cliff”
Annuals of Internal Medicine, 2005.
[4] Wertheimer AI and Santella TM, “Medication Compliance Research,” Journal of Applied Research in Clinical
and Experimental Therapeutics, 2003.
[5] Kohn LT, Corrigan JM, Donaldson MS (Ed.), To Err Is Human: Building a Safer Health System. Compiled by the
Committee on Quality of Health Care in America, The Institute of Medicine, US.
Volume 2, Issue 3, March 2013
Page 368